PROJECT SUMMARY Glutamate (Glu) is the primary excitatory neurotransmitter in the brain, and disruptions of normal glutamate levels are implicated in a variety of major neurological and psychiatric disorders, which has led to major efforts to measure Glu non-invasively. 1H MRS has been exploited to measure Glu but in practice is limited by low sensitivity, low spatial resolution, and partial volume effects. A potential approach to imaging Glu with high sensitivity is to exploit the chemical exchange saturation transfer (CEST) effect between water protons and the rapidly exchanging amine protons of Glu at ? 3 ppm from water. However, although CEST imaging of Glu, named GluCEST, was introduced over 6 years ago and has been successfully applied in diagnosing many preclinical neurological disease models (e.g. tauopathy, dopamine deficiency, Huntington?s disease, and Alzheimer?s disease etc.), it has not been translated to clinical applications at 3 T MRI. This is due to two reasons: First, Glu is in the fast exchange regime and coalesces with water especially at 3 T, which significantly influences GluCEST signals, causing the appearance of false resonances and non-specificities. Although there are many CEST analysis methods that attempt to isolate exchange effects, they are designed only for the slow and intermediate exchange regimes and cannot solve this coalescence effect and fail to quantify GluCEST properly; Second, the molecular origin of GluCEST has not been comprehensively evaluated and its specificity is still under debate. For instance, although Glu has exchangeable amine protons at ? 3 ppm, protein lysine amines have similar chemical shifts and exchange rates in the fast exchange regime, and thus may not be easily distinguished from Glu using CEST. In previous validation of GluCEST, only contributions from the major brain metabolites were considered, but contributions from proteins were ignored. This may be due to that it is difficult in practice to precisely mimic the lysine residues of the wide variety of proteins found in tissues using simple models. This application proposes to overcome those challenges and develop a practical data analysis method to allow routine imaging of Glu. In Aim 1, we will develop and implement a new metric, termed tAREX (tangent theta normalized apparent exchange-dependent relaxation) to address the need for better quantification in the fast exchange limit. Our preliminary analysis and simulations show that tAREX can successfully remove the coalescence effect. In Aim 2, we will use dialysis to remove Glu and other small molecules from samples of brain tissue homogenates to investigate the influence of proteins on GluCEST. Together with measurements on phantoms containing major metabolites, the dialysis of tissue homogenates can provide a comprehensive investigation of the origins of GluCEST. Ultimately, the approach will help future translation of measurements of Glu to clinical scanners and applications. It may be also used in other CEST applications in the fast exchange regime (e.g. PARACEST, Sugar-based CEST, Liposome CEST).